New theoretical techniques are being developed and characterized. These efforts are usually coupled with software development, and involve the systematic testing and evaluation of new ideas. This development is driven by current needs and interests. Specific projects include: - Development of methods for examining reaction mechanism in complex systems. - Unbiased forced sampling of complex conformational transitions and estimation of the potential of mean force along the reaction pathway - Development of the REPLICA/PATH method for determining reaction paths in complex systems using simulated annealing - Development of combined Quantum Mechanical/Molecular Modeling (QM/MM) potentials (Gaussian delocalize MM charges, double link atom method) - GAMESS-UK and CHARMM integration for QM/MM applications - Density functional QM/MM using a double link atom interface - Evaluation of alternate treatments of QM/MM interfaces - Calculation of pK of acids groups using Free Energy Perturbation and PME corrections. - Density functional calculations on the ionization potential of sulfides- Other method development - Development of RMS best-fit restraints; accurate forces, relative restraints - Rational drug design: shape descriptor facility for CHARMM - Development of accurate interaction energy calculations for macromolecules - Evaluation of small molecule/protein binding energy prediction methods - Development of a rapid search strategy for docking two macromolecules Accurate interaction energy calculations for macromolecules using GB/SA (Generalized Born/Surface Area) techniques have been developed and applied to calculating binding energies in solution between protein/DNA complex in water and for peptide-peptide dimerization energies. There has been a continuation of effort towards improving the techniques used to model complex systems with a mixture of quantum mechanics (QM) and classical mechanics (MM). QM/MM methods offer the possibility of treating a region of interest within a biological system quantum mechanically thereby allowing the accurate representation of bond breaking, formation, and electron transfer while also including important structural and charge effects from a surrounding classical region. GAMESS-UK has been tightly integrated into CHARMM to allow studies of catalytic paths in small molecules and enzyme complexes. This extends the QM/MM suite within CHARMM since GAMESS-UK provides DFT (Density Functional Theory). Gaussian convolution (blurring) of classical partial charges has been implemented and tested. These delocalized charges reduce artifacts and improve on the double link atom methodology for treating QM/MM boundary conditions. The interface has been tested on a set of small molecules consisting of a series of alkanes, alcohols, amines, and amino acids, each with a variety of QM methods involving different levels of sophistication. The results indicate that our method and implementation of partioning a molecule into quantum and molecular mechanics regions is robust and would be extremely helpful in studying enzyme mechanisms. A Class A beta-lactamase is being used as a test case for examination of reaction paths. We have also examined the vertical ionization potential of a series of sulfides and disulfides using ab initio and density functional methods. Comparison with available experimental data suggests that B3LYP/6-311+G(d,p) level produces excellent agreement compared to other levels of theory. We propose B3LYP/6-311+G(d,p) should be the level of choice in predicting vertical ionization potential of sulfides for which experimental data is hard to come by.